The Getting Active Project (GAP). The many negative outcomes associated with loneliness in older people have rendered loneliness itself a new public health target. Older adults who feel lonely carry increased risk for reduced quality of life, morbidity, and mortality. The risk of premature mortality related to loneliness is at least as large as the risks arising from such factors as obesity, physical inactivity, alcohol misuse, and smoking. Volunteering is a promising intervention for reducing loneliness in later life. The primary objective of this proposal is to test the hypothesis that a social volunteering program for lonely older adults will lead to reduced loneliness and improved quality of life. National infrastructure for volunteering (The Senior Corps) ensures that volunteering is a highly scalable intervention. We propose to compare the effect of a Senior Corps volunteering intervention versus a self-guided life review active control condition on feelings of loneliness in older adults. Our preliminary data, as well as published studies of volunteering in later life, strongly suggest that volunteering (but not life review) should reduce loneliness. Rigorous experimental study is needed, however, to examine volunteering in both men and women who are lonely, to determine conditions that maximize benefit, and to understand mechanisms. We hypothesize, per tenets of Self-Determination Theory, that increased social engagement and feelings of both usefulness and social support function as psychological mechanisms whereby volunteering reduces loneliness. Understanding these mechanisms will promote effective implementation, allowing communities to adapt volunteering programs while retaining the active ingredients. We will randomly assign older adults (150 women, 150 men) who report loneliness to 12 months of either: 1) a structured social volunteering program providing peer companionship to frail, homebound older adults for at least 4 hours per week, or 2) an active control intervention with self-guided life review. Our aims are as follows: 1) To examine the effect of volunteering on loneliness and quality of life; 2) To examine social engagement, perceived usefulness, and social support as mechanisms for reducing loneliness; 3) To examine conditions under which volunteering is most effective at reducing loneliness. Our intervention is already implemented nation-wide, indicating high feasibility of going to scale (http://www.nationalservice.gov/programs/senior-corps). If effective, volunteering should be ?prescribed? by physicians and promoted by policy. Dissemination and scaling up efforts will involve connecting primary care patients and aging services clients who are lonely with The Senior Corps, which we have shown to be feasible in our companion study, The Senior Connection. Existing infrastructure will make it possible to reach a large proportion of lonely older adults. Reducing loneliness has the potential to improve well-being and save lives. Van Orden